BACKGROUND The relative contribution of risk factors to the development of heart failure remains controversial. Further, whether these contributions have changed over time or differ by sex is unclear. Few population-based studies have been performed. We aimed to estimate the population attributable risk (PAR) associated with key risk factors for heart failure in the community. METHODS Between 1979 and 2002, 962 incident heart failure cases in Olmsted County were age and sex-matched to population-based controls using Rochester Epidemiology Project resources. We determined the frequency of risk factors (coronary heart disease, hypertension, diabetes mellitus, obesity, and smoking), odds ratios, and PAR of each risk factor for heart failure. RESULTS The mean number of risk factors for heart failure per case was 1.9 + or - 1.1 and increased over time (P<.001). Hypertension was the most common (66%), followed by smoking (51%). The prevalence of hypertension, obesity, and smoking increased over time. The risk of heart failure was particularly high for coronary disease and diabetes with odds ratios (95% confidence intervals) of 3.05 (2.36-3.95) and 2.65 (1.98-3.54), respectively. However, the PAR was highest for coronary disease and hypertension; each accounted for 20% of heart failure cases in the population, although coronary disease accounted for the greatest proportion of cases in men (PAR 23%) and hypertension was of greatest importance in women (PAR 28%). CONCLUSIONS Preventing coronary disease and hypertension will have the greatest population impact in preventing heart failure. Sex-targeted prevention strategies might confer additional benefit. However, these relationships can change, underscoring the importance of continued surveillance of heart failure.